Interleukin‐17 and ischaemic stroke

IL‐17 family was conducted with 6 members, most of them are widely involved in a variety of acute and chronic inflammatory responses, especially IL‐17A. After stroke, cerebral ischaemia and hypoxia lead nerve cell necrosis and release a large amount of damage‐associated molecular patterns and inflammation factors to activate IL‐17A. Then, IL‐17A promotes the development of stroke by inducing the secretion of inflammatory factors (such as TNF‐α, IL‐6, CXCL1), recruiting neutrophils to infiltrate into central nervous system and impairing the integrity of blood–brain barrier.

Summary

Interleukin‐17 (IL‐17) is a cytokine family that includes 6 members, IL‐17A through IL‐17F, most of them are reported to have pro‐inflammatory role. Through binding to their receptors (IL‐17Rs), IL‐17 activates the intracellular signalling pathways to play an important role in autoimmune diseases, including rheumatoid arthritis (RA) and multiple sclerosis (MS). Ischaemic stroke is a complex pathophysiological process mainly caused by regional cerebral ischaemia. Inflammatory factors contribute to the physiological process of stroke that leads to poor prognosis. IL‐17 plays a crucial role in promoting inflammatory response and inducing secondary injury in post‐stroke. Though immune cells and inflammatory factors have been reported to be involved in the damage of stroke, the functions of IL‐17 in this process need to be elucidated. This review focuses on the pathological modulation and the mechanism of IL‐17 family in ischaemic stroke and seeking to provide new insights for future therapies.

Abbreviations

AP‐1

activator protein 1

BBB

blood–brain barrier

Bcl‐2

B‐cell lymphoma‐2

BDNF

brain‐derived neurotrophic factor

Blk

B‐lymphoid tyrosine kinase

BMEC

brain microvascular endothelial cell

CNS

central nervous system

CREB

cAMP‐response element‐binding protein

CTLA8

T‐lymphocyte‐associated antigen 8

DAMP

damage‐associated molecular pattern

DC

dendritic cell

EAE

experimental autoimmune encephalitis

ERK

extracellular signal‐related kinase

Foxp3

forkhead box P3

GSK‐3β

glycogen synthase kinase‐3β

HIF‐1α

hypoxia‐inducible factor‐1α

HuR

human antigen R

IKK

inhibitor of kappa‐B kinase

INF‐γ

interferon‐gamma

JAK1

Janus kinase 1

JNK

Janus kinase

Lef1

lymphoid enhancer factor 1

MKK

mitogen‐activated protein kinase kinases

MLC

myosin light chain

MS

multiple sclerosis

NDR1

nuclear Dbf2‐related kinase 1

NET

neutrophil extracellular trap

NF‐κB

nuclear factor kappa‐B

PI3K/AKT

phosphatidylinositol 3 kinase/protein kinase B

pMCAO

permanent middle cerebral artery occlusion

RA

rheumatoid arthritis

RORγt

retinoid‐related orphan receptor gamma‐t

SEFIR

SEFIR: SEF/IL‐17 receptor

SF2

splicing factor 2

STAT3

signal transducer and activator of transcription 3

SYK

spleen tyrosine kinase

TAB2/3

TGF‐β‐activated kinase 1/MAP 3K7 binding protein 2/3

TAK1

TGF‐β activated kinase 1

TJ

tight junction

tMCAO

transient middle cerebral artery occlusion

TRAF

tumour necrosis factor receptor‐associated factor

TRPC6

transient receptor potential cation channel 6

TYK2

tyrosine kinase 2

MAPK

mitogen‐activated protein kinase

INTRODUCTION

The incidence, relapse and mortality rates of stroke are high around the world. From 2006 to 2016, stroke happens more frequently with advancing age in both males and females, and the number of strokes is predicted to increase by over 20% compared with 2012 in the United States by 2030. ¹ A recent report from American Heart Association/American Stroke Association demonstrated the age‐adjusted stroke mortality decreased in comparison with previous years, regardless of gender, whereas the actual number of stroke deaths increased by almost 4%. ¹ In 2016, stroke was accounted for 42.2% of deaths of all neurological disorders. The substantial high fatality rate makes it a serious threat to public health all over the world. ² It is noteworthy that ischaemic stroke caused by regional cerebral blood supply disorders makes up 87% of all stroke patients. ¹

The pathophysiological mechanism of ischaemic stroke is complex; various cellular, molecular and signalling pathways are involved in the process of stroke. The intricate process of stroke includes bioenergy failure, imbalance of ion homeostasis, apoptosis, excitotoxicity, oxidative stress, immune reaction, inflammation, etc. ³ , ⁴ The physical factors and chemical factors of anastomosis cause massive neurodegeneration. After cerebral infarction occlusion, the anoxic and malnourished nerve cells involving neurons, glial cells were necrotic. Necrotic nerve cells release excessive damage‐associated molecular patterns (DAMPs), as endogenous alarms, including high mobility group protein B1 (HMGB1), reactive oxygen species (ROS), peroxiredoxins (PRXs) and heat‐shock proteins (HSPs). Many DAMPs are sensed by Toll‐like receptors (TLRs) and receptor of advanced glycation end product (RAGE), both of which are the sensor of the innate immune system. ⁵ After combining with the TLRs and RAGE, DAMPs stimulate several inflammatory signalling pathways in the innate immune cells such as microglia and brain microvascular endothelial cells (BMECs), then release inflammatory mediators and thus cause central inflammatory events. ⁶ Notably, the up‐regulation of matrix metalloproteinases (MMPs), most of which are expressed at the undetected level in the normal model, shows the harmful effect of impairing the integrity of blood–brain barrier (BBB). Damaged BBB facilitates the transfer of DAMPs into peripheral blood and activates immune cells (including neutrophils, macrophages, and T and B lymphocytes) to migrate and infiltrate into cerebral parenchyma, finally leading to a robust activation of the immune system. ⁵ , ⁶ Studies have shown that 70–80% of nerve damage is related to post‐stroke inflammatory response. ⁷ Therefore, tempering the inflammatory response is essential for improving the prognosis of ischaemic stroke.

It has been widely accepted that the inhibition of lymphocyte activation and infiltration significantly improves the pathological outcome of a stroke. ⁸ , ⁹ In T‐ and B‐cell‐deficient mice, inflammatory factors, including TNF‐α and IL‐6, were significantly reduced with ameliorating infarct area after ischaemia. ¹⁰ However, reconstitution of B lymphocytes could not reverse this destructive change. ¹¹ These results suggested that T cells might be more critical in modulating stroke damage.

T lymphocytes can be divided into αβ T cells and γδ T cells basing on the types of T‐cell receptor (TCR) on the cell membrane. Furthermore, depending on the surface molecules, αβ T cells can be divided into CD4⁺ T cells and CD8⁺ T cells; the former can be further divided into T helpers (Th) and regulatory T cells (Treg). ¹² Notably, all types of T cells are closely involved in triggering inflammatory events in stroke. ¹³ In the acute phase of ischaemic stroke, Th1 and Th17 were destructive. ¹⁴ , ¹⁵ Since the 20th century, great progress has been made in the study of γδ T cells in stroke. CD4⁻ IL‐17‐secreting γδ T cells (IL‐17⁺ γδ T cells), Th1 and Th17 were considered as harmful responders in sites of ischaemic hemisphere injury. ¹⁴

As previously reported, IL‐17⁺ γδ T cells, the principal recruited immune cells from the intestine, have been documented as a major source of neuroinflammation cascades in acute ischaemic brain damage. ¹⁶ , ¹⁷ According to Corinne et al., imbalance of intestinal flora homeostasis promoted the migration of IL‐17⁺ γδ T cells to meninges by altering the activity of dendritic cell, which therefore aggravated ischaemic brain injury by producing IL‐17 and its following by‐products, including CXCL1 and CXCL2. ¹⁶ IL‐17 is expressed by both Th17 and γδ T cells and is detrimental to stroke. ¹⁸ , ¹⁹

This review will focus on discussing the mechanism of IL‐17 in ischaemic stroke to provide a theoretical framework for the possibilities of targeting IL‐17 for the treatment of ischaemic stroke.

THE IL‐17 AND IL‐17 RECEPTOR FAMILY

In 1993, a mouse T‐cell hybridoma, cytotoxic T‐lymphocyte‐associated antigen 8 (CTLA8), was first cloned and then found to play an unusual role in the immune response. ²⁰ In 1995, a nucleotide sequence related to mRNA instability was found in the 3ʹ untranslated region (3ʹ‐UTR) of CTLA8. Although the strand shares great similarities with most interleukins, CTLA8 was renamed as IL‐17. ²¹ In the same year, the human IL‐17 cDNA sequence was cloned from the CD4⁺ T cells, and it was found to exhibit 63% identity to the gene open reading frame of mouse CTLA8. This finding further confirmed the previous conclusion. ²² Since then, the five other cytokines of IL‐17 family, including IL‐17B, C, D, E and IL‐17F, were cloned (Fig. 1). Among them, IL‐17F has the most sequence identity with IL‐17A (50%), IL‐17E (also known as IL‐25) shares 17%, and IL‐17B, IL‐17D and IL‐17C share 29%, 25% and 23% identities with IL‐17A protein, respectively. ²³ , ²⁴ , ²⁵ , ²⁶ The above six cytokines form the IL‐17 family due to their similarities in structure.

Figure 1.

IL‐17 family of cytokines was found. A timeline showing the discovery of IL‐17 and its receptor family. References in 18, 19, 21, 22, 23, 24.

IL‐17 family activates downstream molecules by binding to the specific receptor complexes on the cellular membrane. The IL‐17 receptor (IL‐17R) family consists of five subunits: IL‐17RA to IL‐17RE. ²⁷ Although the molecular weight of IL‐17R is large (about 860 amino acids), there is no evidence to show IL‐17R family is similar to other cytokine receptors. ²⁷ The subunits of the IL‐17R family contain conserved functional domains, including the extracellular fibronectin III‐like domain and the intracellular signal transduction via motif “SEFIR” domain. ²⁸ By binding to IL‐17R, IL‐17 mediates inflammatory responses via NF‐κB and MAPK/AP‐1 pathways. ²⁹ (see below)

In the past 20 years, studies have confirmed that IL‐17A plays a vital role in infectious and non‐infectious diseases such as chronic hepatitis B virus, autoimmune diseases, cancer and ischaemic stroke. ⁴ , ¹² , ³⁰ In addition, CD4⁺ T‐cell subsets that secrete IL‐17A were named as Th17 cells. ³¹ Since then, other innate or adaptive immune cells including γδ T cells, natural killer T (NKT) cells, natural killer (NK) cells, group 3 innate lymphoid cells (ILC3s) and neutrophils have been found to also secrete IL‐17A. ³² The IL‐17A‐secreting cells were found to express CCL20 receptor (CCR6) and IL‐23 receptor (IL‐23R). ³² This finding suggested that chemokine CCL20 and IL‐23 may be important in the function of IL‐17A‐expressing cells.

In 2001, IL‐17F was successfully cloned. ²⁶ Due to its high amino acid sequence identity with IL‐17A, ³³ they were found to make up a heterodimer (IL‐17A/F), and share the same receptor (IL‐17RA/IL‐17RC heterodimer). ³⁴ In addition, they show similar biological effects in many diseases, including psoriasis ³³ and asthma. ³⁵ IL‐17A has high affinity towards IL‐17RA expressed in haematopoietic cells. However, IL‐17F, which is secreted by Th17 cells, γδ T cells, monocytes, mast cells and basophils, has a high affinity towards IL‐17RC expressed on non‐haematopoietic cells. ³⁶ This phenomenon determines the functional differences between IL‐17A and IL‐17F. ³⁷ Studies have confirmed that IL‐17F plays an important role in inflammatory bowel disease (Crohn's disease and ulcerative colitis), ³⁷ but whether IL‐17F plays a role in ischaemic stroke remains unknown.

IL‐17A/IL‐17R SIGNALLING PATHWAYS

IL‐17R, a constitutive heterodimer composed of two subunits IL‐17RA and IL‐17RC, is the receptor of IL‐17A/A, IL‐17A/F and IL‐17F/F. ³⁸ IL‐17RC shares 23% amino acid sequence identity with IL‐17RA, both of which have a self‐conservation cytoplasmic SEFIR domain. ³⁸ The key transcriptional factor in IL‐17A signalling is nuclear factor kappa‐B activator 1 (Act1). ³⁹ Act1 is a protein with a SEFIR domain at the C‐terminal, and a tumour necrosis factor receptor‐associated factor (TRAF) family domain at the N‐terminal. When binding to IL‐17R, IL‐17A recruits and combines with Act1’s SEFIR domain, then recruits and ubiquitinates members of the TRAF family for through U‐box domain. ²⁹ , ⁴⁰ Ubiquitinated TRAF activate, TGF‐β‐activated kinase 1 (TAK1) that activates the complex of IKK (IKKα, IKKβ, IKKγ) ⁴¹ to ubiquitinate the two specific serine sequences of IκBα. Consequently, NF‐κB dimers dissociate from the complex. NF‐κB dimers translocate into the nucleus and bind to the corresponding target genes, finally promote the gene transcription of pro‐inflammatory cytokines. ⁴¹ , ⁴² In addition, TRAF6/TAK1 also promotes the transcription of IL‐17A target genes by activating MAPK/AP‐1 pathway ⁴³ and C/EBP pathway ⁴⁴ . However, extracellular signal‐related kinase (ERK) and glycogen synthase kinase‐3β (GSK‐3β) could phosphorylate the Thr188 and Thr179 of C/EBPβ, which in turn negatively regulate the expression of IL‐17A target genes ⁴⁴ (Fig. 2).

Figure 2.

The IL‐17A signalling pathway. IL‐17 binds to its receptor, which is composed of IL17RA and IL‐17RC. Upon ligand binding, activated Act1 phosphorylated TRAF6 and triggers the transcription of TRAF6‐dependent target genes such as NF‐κB, CEBP and MAPK/AP‐1. The IL17A signalling pathway is regulated by many molecules, including TRAF family, HuR and so on. TRAF2‐TRAF5 complex and HuR stabilize the mRNA of IL‐17A target genes and contribute to the pathogenesis of stroke, while TRAF3 and TRAF4 seem to play a negative role. TRAF3 competes with Act1 to IL‐17R, while TRAF4 competes with TRAF6 for Act1. Besides, ERK1/2 and GSK‐3β phosphorylate the Thr188 and Thr179 of C/EBPβ, in turn, inhibits of CEBP pathway. References in 28, 41, 42, 44.

IL‐17A MODULATION IN STROKE

According to Tables 1 and 2, IL‐17A was elevated in brain tissue and peripheral blood after acute ischaemic brain injury in both experimental models and patients, and the level of IL‐17A was much higher than the level in recovery phases. Almost all the experimental data confirmed that IL‐17A was elevated within one day and peaked on the third day and then dropped a little bit on the following days after ischaemia injury in the mouse model. However, the actual situation of patients is different from that of experimental models. First of all, the patients’ data were usually taken from interval values, which is hard to track to a specific day. Secondly, it is not always clear whether there is ischaemia–reperfusion in the patients. Thirdly, the diagnostic criteria of stroke patients in different hospitals or different years are slightly different. Lastly, most of the patients are the elderly, who may have basic diseases such as hypertension or atherosclerosis, so individual differences are also relatively large. As a result, the experimental model in vivo is not always appropriate for meeting human conditions. The differences between experimental models and patients might be related to the following reasons: (1) species and genetic differences; (2) variable inclusion criteria; (3) individual differences among the elderly patients; and (4) data availability differences. Nevertheless, there is no doubt that the increased IL‐17A in brain tissue, serum and peripheral immune organs during acute stage of ischaemic stroke are related to brain injury in both experimental models and patients. (Tables 1 and 2)

Table 1.

IL‐17 was elevated in peripheral blood in ischaemic stroke patients and elevated in ischaemic hemisphere in patients died of ischaemic stroke

Objects	Inclusion criteria					Detected organs or tissue	Analysed time‐points	Clinical characteristic	Releasing cells	Pub. year	Ref.
	Sex	Age range	Type of stroke	Numerosity	Diagnosis criterion
Patients	18 females, 11 males	49–87	Ischaemic stroke	29 patients, 18 healthy individuals	Cerebral disturbance of clinical symptoms over 24 hours or recovery within 24 hours.	PBMCs	IL‐17 mRNA elevated more from day 1 to day 3 than from day 20 to day 31 after onset of symptoms.	–	–	1999	45
	–	–	Ischaemic stroke	26	Patients died of ischaemic stroke.	Ischaemic hemisphere	IL‐17 was elevated on day 2 and peaked on day 3 to 5.	–	–	2000	46
	28 females, 22 males	45–76	Ischaemic stroke	50 patients, 30 healthy individuals	World Health Organization criteria.	Plasma	IL‐17A was elevated on day 7 significantly and remained at a low level on day 28.	Hypertension, diabetes, hyperlipidaemia	Th17 cells and γδ T cells	2014	47
	19 females, 11 males	66–74	Ischaemic stroke	30 patients, 30 healthy individuals	World Health Organization criteria.	PBMCs	IL‐17A was elevated on day 1, 5, 10 and peaked on day 1 after stroke onset.	–	Th17 cells	2018	48
	300	–	Ischaemic stroke	300 patients, 300 healthy individuals	Ischaemic stroke diagnosed, regardless of gender and age.	Serum	IL‐17 was elevated after stroke onset.	–	–	2019	49
	26 females, 78 males	18–80	Ischaemic stroke	104 patients, 90 healthy individuals	Ischaemic stroke diagnosed within 7 days along with large‐artery atherosclerosis	–	–	Hypertension, diabetes mellitus, glucose, white blood cell count, high‐density lipoprotein	–	2019	50

PBMCs, peripheral blood mononuclear cells; Pub. year, published year.

Table 2.

IL‐17 was elevated in experimental models

Objects	Sex and numerosity	Age/weight	Type of mouse	Modelling	Detected parts	Analysed time‐points and molecular results	Releasing cells	Pub. year	Ref.
Animals	48 males	250–300 g	SD rats	pMCAO	Brain tissue, blood, spleen, lymph node	IL‐17 mRNA was elevated at 1 h in the brain, blood, spleen and lymph node and then peaked on the sixth day, 6, 12 and 6 h, respectively. Moreover, INF‐γ, IL‐8, IP‐10, MIP‐2 and IL‐4 were up‐regulated and IL‐10 was down‐regulated in brain after pMCAO.	–	2001	51
	72 males	250–300 g	SD rats	pMCAO	Ischaemic hemisphere	IL‐17 mRNA was elevated at 1 h and remained at a higher and then peaked on day 6. CD3+ was also elevated at 1 h and then peaked on day 6 in the brain.	T cells	2005	46
	Male	7–17 weeks ole	C57BL/6 mice	MCAO/R	Ischaemic brain tissue	IL‐17 was elevated on day 3 followed by IL‐23, which was elevated on day 1 after MCAO‐R. Moreover, most of IL‐17‐producing cells were γδ T cells.	γδ T cells	2009	17
	‐	–	C57BL/6 mice	tMCAO	Ischaemic hemisphere	T cells were detectable at 1 h after tMCAO and then peaked on day 3, while IL‐17+ γδ T cells were found at 6 h and peaked at day 3. Moreover, CXCL‐1 and MMP3 were activated by IL‐17A.	γδ T cells	2012	52
	48 males	25–30 g	C57BL/6 mice	pMCAO	Ischaemic brain tissue, serum	Both IL‐23 and IL‐17 were elevated on day 6 in brain tissue and serum.	–	2013	53
	Male	8–10 weeks old	C57BL/6 mice	tMCAO	Penumbra tissue	HMGB1 and IL‐17A were elevated on day 3 after tMCAO. IL‐17A knockout reduced the damage of brain, while rIL‐17A increased, but rHMGB1 failed to aggravate the injury.	γδ T cells	2014	54
	Male	8–10 weeks old	C57BL/6 mice	tMCAO	Penumbra tissue	IL‐17A was elevated on day 1, peaked on day 3 and remained slightly decreased on day 6 in the hemisphere penumbra. IL‐17A activated TRPC6 channels on day 3 to aggravate the brain damage.	‐	2014	55
	33 Males	25 g	C57BL/6 mice	pMCAO	Ischaemic brain tissue	IL‐17 was elevated on day 5 after ischaemic injury, while IL‐23p19 knockout alleviated the damage. Moreover, IL‐23p19 knockout increased the expression of INF‐γ and Foxp3.	Th17 cells and γδ T cells	2015	56
	Male	250–300 g	SD rats	tMCAO	Ischaemic hemisphere, peripheral blood	IL‐17 was elevated on days 1 and 3 after modelling. IL‐10 and TGF‐β were decreased on day 3, while IL‐1β and TNF‐α were increased on day 1 in brain. Treg cells were detected on day 3, while γδ T cells were increased on day 1 after cerebral ischaemic in peripheral blood.	γδ T cells	2016	57
	Male	–	C57BL/6 mice	tMCAO	Ischaemic hemisphere	γδ T cells were infiltrated at 12 h and kept rising on day 3. Most of them were Vγ6+/CCR6+ γδ T cells that produced IL‐17, not INF‐γ to aggravate the brain injury.	Vγ6+ γδ T cells	2017	58
	Male	8–10 weeks old	C57BL/6J mice	MCAO/R	CSF, serum, peri‐infarct homogenates	IL‐17A was elevated at 12 h in peri‐infarct homogenates and CSF, while elevated on day 1 and peaked on day 3 in serum after MCAO/R.	Astrocytes	2017	59
	Male	–	C57BL/6 mice	tMCAO	Peripheral blood, cerebral hemisphere	IL‐17 was elevated on day 2 and peaked on day 3, then slightly decreased on day 5 in peripheral blood after tMCAO. IL‐17RA was slightly increased though 12 h to 5 days with no differences in cerebral.	–	2018	60
	20 males	6 weeks old	C57BL/6 mice	MCAO	Intestinal intraepithelial lymphocyte, spleen	IL‐17+ γδ T cells were up‐regulated on the third day in both small intestine and spleen, while CD4+ CD25+ T cells were down‐regulated in spleen and CD4+ Foxp3+ Treg cells were down‐regulated in small intestine.	γδ T cells	2019	50

C57BL/6 mice, C57BL/6 background mice; Foxp3, forkhead box P3; INF‐γ, interferon‐gamma; MCAO, middle cerebral artery occlusion; MCAO/R, MCAO/reperfusion; pMCAO, permanent MCAO; rHMGB1, repletion of exogenous HMGB1; rIL‐17A, repletion of exogenous IL‐17A; SD rats, Sprague–Dawley rats; tMCAO, transient MCAO.

Previous studies mainly focused on the acute phase of stroke; however, more studies confirmed that IL‐17A might dampen specific stroke pathophysiological consequences in sub‐acute and convalescence stage. According to Lin et al., ⁶¹ IL‐17A played a biphasic role on different time‐points after ischaemic stroke in mice. IL‐17A‐releasing γδ T cells peaked on day 3 and aggravated the ischaemic injury. However, IL‐17A‐releasing astrocytes peaked on day 28, which seems to alleviate the damage by promoting neurogenesis and synaptogenesis, maintaining the neuronal differentiation and the survival of subventricular zone neural precursor cells via activated P38 MAPK/calpain 1 signalling. In this review, we still focused on the acute phase of stroke to study the possible mechanism of IL‐17A injury.

IL‐17A PROMOTES NEUTROPHIL INFILTRATION IN STROKE

IL‐17R is expressed on both glial cells and BMECs and up‐regulated after stroke in experimental models. ⁵² , ⁶² By binding with IL‐17R, IL‐17A promotes glia and BMECs to secrete large amounts of CXCL1 ⁵² , ⁶⁰ , ⁶² to induce neutrophils to infiltrate the cerebral parenchyma and inflict damage in stroke. ⁶³ , ⁶⁴ The main manifestations of neutrophils effects include the following: (1) activate MMPs (especially MMP‐2, ‐9) and hydrolyse endothelial cell tight junction proteins (TJs) to destroy the structural integrity of the BBB ⁶⁵ , ⁶⁶ ; and (2) secrete inflammatory cytokines, chemokines and adhesion molecules, and induce the infiltration of CD8⁺ T cells via CXCL12. ⁶⁵ Infiltration of CD8⁺ T cells could induce the lysis and apoptosis of nerve cells through perforin pathway, granular enzyme pathway and Fas/FasL pathway; ¹² (3) promote the formation of thrombosis. ⁶⁷ Neutrophils not only adhere to endothelial cells through the interaction between leucocyte function‐associated antigen (LFA) and ICAM‐1 (thereby activating platelets and promoting thrombosis), ⁶⁷ but also promote the formation of neutrophil extracellular trap (NET) through the release of DNA, histone and specific granule proteins. NET can capture fragments and dead cells to activate the exogenous coagulation pathway to form thrombus. ⁶⁷ All the damaging effects lead to "no reflow" in the ischaemic area and aggravate the disturbance of downstream perfusion, leading to brain oedema, post‐embolization haemorrhage, larger infarction and worse outcome. ⁶³ , ⁶⁸ Notably, the expression of CXCL1, the main cytokine of neutrophils chemotaxis, depends on the induction of IL‐17A. ⁵² The blockage of IL‐17A signalling (in Il17ra ^−/−mouse model or with anti‐IL‐17A treatment) significantly inhibits neutrophil infiltration and reduces infarct size ⁵² (Fig. 3).

Figure 3.

Production and function of IL‐17 after stroke. After stroke, cerebral ischaemia and hypoxia lead nerve cells necrosis and release a large amount of DAMPs and inflammation factors. When DAMPs are combined with TLRs on DCs, it promotes the DC‐derived IL‐23 releasing from DCs and then IL‐23 promotes the release of IL‐17 by γδ T cells, Th17 cells NKT cells and so on. Finally, IL‐17 leads to poor prognosis of stroke in three ways: (1) inducing the secretion of pro‐inflammatory factors; (2) recruiting neutrophils to infiltrate into CNS; (3) impairing the integrity of BBB.

IL‐17A IMPAIRS BBB INTEGRITY

BBB is an important component of the neurovascular unit (NVU) and is the first barrier to maintain the homeostasis of the microenvironment in the brain. ⁶⁹ Destroyed BBB cohesiveness will cause a devastating effect on nerve cells’ survival and function. The mechanism of how IL‐17A impairs the cohesion of BBB is complex. Firstly, it down‐regulates the expression of tight junction (TJ) molecules such as occludin and claudin‐5 ⁷⁰ and promotes the elevation of MMP‐2 and MMP‐9 in BMECs to hydrolyse TJs. ⁷¹ As previously described, IL‐17A plays a powerful role in activating chemokine CXCL1 to induce neutrophils to infiltrate the cerebral parenchyma. Moreover, neutrophils produced and activated a large number of MMPs, such as MMP2, MMP3 and MMP9. MMPs destroy the integrity of BBB by hydrolysing TJs and improving BBB permeability. ⁵² , ⁶⁶ Secondly, it activates BMECs to release reactive oxygen species (ROS) to induce oxidative stress and activate myosin light chain (MLC). ⁷⁰ Phosphorylated MLC interacts with cytoskeletal actin and induces the contraction of BMECs and eventually leads to the enlarged gaps between endothelial cells. ⁷⁰ , ⁷² IL‐17A induces inflammatory response by up‐regulating ICAM‐1 and IL‐6 in BMECs. ⁷⁰ It consequently induces endothelial apoptosis through mitochondrial apoptosis pathway ⁷³ and then results in the destruction of BBB. Experiments in vitro showed that IL‐17A up‐regulates the expression of IL‐17R in brain endothelial cells, while anti‐IL‐17A effectively protects the cohesion of BBB ⁷¹ (Fig. 3).

IL‐17A INDUCES NEURONAL APOPTOSIS

With current knowledge, transient receptor potential cation channel 6 (TRPC6) can phosphorylate cAMP‐response element‐binding protein (CREB), which in turn activates brain‐derived neurotrophic factor (BDNF) and anti‐apoptotic protein B‐cell lymphoma‐2 (Bcl‐2). ⁷⁴ As a pro‐survival factor for neuron, CREB plays an important role in promoting both neuron survival and nutritional support. ⁷⁵ TRPC6/CREB pathway is important to maintain nerve cell's survival and function after stroke by improving hypoxia tolerance of nerve cells. ⁷⁶ , ⁷⁷ Du et al. showed that inhibition of TRPC6/CREB pathway leads to reduce the expression of BDNF and Bcl‐2, as well as significantly reduced the hypoxia tolerance of neurons. ⁷⁶ However, inhibiting TRPC6 degrades the defensive effect of ischaemic brain damage. After stroke, IL‐17A increases rapidly and activates calpain to mediate the TRPC6 hydrolysis. This process reduces the expression of BDNF and Bcl‐2 and results in neuron death. ⁵⁵ It is noteworthy that IL‐17A promotes the expression of apoptosis‐related proteins including caspase‐3, caspase‐9 and Bax and increases the ratio of Bax/Bcl‐2 after brain injury, thus initiates the process of neuronal apoptosis ⁷⁸ (Fig. 3).

SYNERGISTIC EFFECTS OF IL‐17A WITH IL‐6 AND IL‐22

Although IL‐17A is a mild stimulator of inflammatory response on its own, it activates the inflammatory signalling pathways of C/EBP, NF‐κB and MAPK that can promote expression of pro‐inflammatory cytokines and amplifying the inflammatory response. ³² It has been reported that the expression of C/EBP δ and NF‐κB is up‐regulated in glial cells, BMECs and macrophages after ischaemia and hypoxia, which is closely related to the activation of IL‐17A pathway. ⁴⁴ , ⁷⁹ After the activation of IL‐17R/Act1, TRAF6 is recruited and activates downstream C/EBP β, C/EBP δ and NF‐κB, ⁴⁴ , ⁷⁹ all of which are key proteins of IL‐6 transcription. When C/EBP β or C/EBP δ binds to the IL‐6 promoter, they induce macrophages, BMECs and fibroblasts to up‐regulate the expression of IL‐6, thereby amplify the inflammatory response.

IL‐22, a member of IL‐10 family, like IL‐17A, is secreted by CD4⁺ αβ T cells, γδ T cells, NKT cells and ILC cells. ⁸⁰ However, the pro‐inflammatory effects of IL‐22 are weaker than IL‐17A. IL‐22 binds to the receptor (IL‐10Rβ/IL‐22R) and transmits downstream signals through Janus kinase 1 (JAK1) and TYK2. After phosphorylation of STAT3, it directly transmits into the nucleus to regulate the transcription of target genes, either AKT/mTOR or MAPK signalling pathways can be activated as well. ⁸¹ Ultimately, the expression of cytokines (IL‐6, G‐CSF) and chemokines (CXCL1, CXCL5, CXCL9) can be activated. ⁸² IL‐17A and IL‐22 improve their pro‐inflammatory effects by a synergistic effect. This synergistic effect has been well proved in asthma, airway inflammation, cancer and other autoimmune diseases. ⁸⁰ , ⁸³ Although there is no direct evidence showing IL‐17 and IL‐22 are synergistic in promoting stroke, the expression of both IL‐17 and IL‐22 are increased in post‐stroke might provide some counter‐evidence. ⁸⁴ , ⁸⁵ Therefore, the possible synergies between IL‐17 and IL‐22 in stroke need further investigation.

IL‐23/IL‐17A AXIS AND ISCHAEMIC STROKE

Both innate and adaptive immunity are involved and interacted with each other in ischaemic stroke. ⁸⁶ After ischaemia, innate‐like lymphocytes are activated rapidly. Innate immunity triggers the first step of inflammatory cascade and then activates the adaptive immunity by presenting antigens. ⁸⁶ Remarkably, the IL‐23/IL‐17 axis is mostly a bridging mechanism between innate immunity and adaptive immunity after stroke. ¹⁸ , ⁸⁷

Dendritic cells (DCs) induce innate immune response in infarct area and rapidly spread to the whole body. This immune response occurs before the adaptive immune response and does not mediate antigen presentation. ⁸⁸ Of note, the combination of DAMPs released by hypoxic neurons and TLRs can induce this immune response. ¹⁸ Then, DCs rapidly release a large number of cytokines and chemokines, especially DC‐derived IL‐23, ⁸⁷ a heterodimer composed of p19 and p40 subunits. IL‐23R is highly expressed in mature Th17 cells and γδ T cells, but not in naïve T cell. ³² As a result, it is generally believed that IL‐23 can amplify the secretion of IL‐17A, while it is not enough to induce the T‐cell differentiation ⁸⁹ (Fig. 3). The key combination that can effectively induce Th17 cell differentiation is TGF‐β+IL‐6, ⁹⁰ , ⁹¹ TGF‐β+IL‐21 ⁹¹ , ⁹² or IL‐23+IL‐6+IL‐1β. ⁹³

When IL‐23 binds to IL‐23R on Th17 cells and γδ T cells, it activates the classic tyrosine‐dependent pathway and phosphorylates STAT3 through JAK2/STAT3 and PI3K/AKT signalling pathways. ⁹⁴ What's more, it activates STAT3 by non‐tyrosine‐dependent pathway. ⁹⁵ pSTAT3 then activates the downstream proteins of transcription factor c‐Maf, a member of the activator protein 1 (AP‐1) family, and then produces a large number of IL‐17 through four mechanisms. Firstly, STAT3 directly activates the nuclear transcription factor retinoid‐related orphan receptor gamma‐t (RORγt) of Th17 cells. ⁹⁶ Secondly, STAT3 directly transmits into the nucleus and binds to the promoters of Il17a, Il17f, Il21 to promote the expression of IL‐17A. ⁹⁶ Thirdly, STAT3 activates RORγt in a RORγt‐dependent manner through c‐Maf. ⁹⁷ , ⁹⁸ Lastly, STAT3 induces IL‐17A secretion in a RORγt‐independent manner through c‐Maf. ⁹⁵

As IL‐23 is a powerful stimulant of IL‐17A, the concept of IL‐23/IL‐17 axis has been proposed and applied to a variety of autoimmune and inflammatory diseases, including ischaemic stroke. ⁸⁷ , ⁹⁹ After stroke, the expression of IL‐23 and IL‐17 in both serum and ischaemic brain tissue is significantly increased, ⁵³ , ⁵⁶ which eventually leads to the destruction of BBB, neuroinflammation and neuron death. Inhibition of IL‐23 (knockdown IL‐23p19) ⁵⁶ or anti‐IL‐17 ⁷¹ can significantly reduce the cerebral ischaemia as well as improve neurological behaviour. IL‐23/IL‐17 axis as an important pathway for post‐stroke inflammation is also plausible. Various targets are involved in the IL‐23/IL‐17 axis. Yet, few studies are on the roles for IL‐23/IL‐17 in stroke, which warrants further investigation as they could be important targets for stroke treatment.

REGULATION OF IL‐17A EXPRESSION AFTER STROKE

Th17 cells and γδ T cells infiltrate into the ischaemic area and serve as the main sources of IL‐17A during post‐stroke. ¹⁷ RORγt is a key transcription factor for IL‐17 expression in Th17 cells and γδ T cells. ¹⁰⁰ The transcription and translation of RORγt are regulated by many molecules.

After stroke, the expression of hypoxia‐inducible factor‐1α (HIF‐1α) is up‐regulated. Although HIF‐1α usually acts as a neuroprotective factor for its role in angiogenesis, ¹⁰¹ glycolysis and glucose transport, ¹⁰¹ and neurogenesis, ¹⁰² the role of HIF‐1α in stroke is complex. Overexpression of HIF‐1α promotes neuronal inflammation and neuronal apoptosis. ¹⁰³ HIF‐1α directly activates RORγt expression by binding to RORC promoter, in order for IL‐17A expression. ¹⁰⁴ It is worth noting that histone acetyltransferase p300 recruited by HIF‐1α can greatly promote Th17 cells to secrete IL‐17A. ¹⁰⁴ Recently, c‐Maf has been identified as a key transcription factor in γδ T ⁹⁸ and Th17 cells. ¹⁰⁵ As mentioned above, c‐Maf plays a key role in promoting the transcription of RORγt. Therefore, it promotes the expression of IL‐17A by up‐regulating RORγt, B‐lymphoid tyrosine kinase (Blk) and spleen tyrosine kinase (SYK), as well as inhibiting lymphoid enhancer factor 1 (Lef1). ⁹⁸ In this manner, RORγt is important to the IL‐17A expression. Additional mechanisms likely exist to contribute to the complex regulation of RORγt; further research is needed to reveal new targets for the treatment of stroke.

REGULATION OF IL‐17R SIGNALLING AFTER STROKE

TRAF family, TRAF2 to TRAF6, is widely involved in multiple pathways, including IL‐17A signalling. ¹⁰⁶ TRAF6 is involved in the inflammatory response of the central nervous system (CNS). Liu et al. ¹⁰⁷ demonstrated that TRAF6 was significantly increased in rats with pMCAO model. As a key modulator in IL‐17A signalling pathway, TRAF6 positively regulates the expression of IL‐17A target genes. ⁴² TRAF2/5 maintains the stability of mRNA. ¹⁰⁶ Act1 recruits and binds TRAF2/5 at Ser311 site and forms Act1‐TRAF2/5‐ASF complex. This complex blocks the binding of splicing factor alternative splicing factor (ASF) to mRNA 3'UTR region to prolong the half‐life period of chemokine mRNA. ¹⁰⁶ , ¹⁰⁸ TRAF2/5 recruits and phosphorylates human antigen R (HuR), and mRNA splicing factor 2 (SF2) competitively binds to RNA‐binding protein (RBP) in the 3'UTR region. Thereby, the binding stabilizes TNF‐α mRNA, IL‐6 mRNA, CXCL‐1 mRNA and ICAM‐1 mRNA to amplify the inflammatory effects. ⁴¹ , ⁴² , ¹⁰⁹ HuR was reported to increase significantly in brain tissue of mice with ischaemia–reperfusion injury after tMCAO. ¹¹⁰

While TRAF6/2/5 promotes the expression of IL‐17A, TRAF3/4 inhibits IL‐17A signal transduction. TRAF3 competes with Act1 to IL‐17R, ¹¹¹ while TRAF4 and TRAF6 competitively bind TB binding sites on Act1. ¹¹² Thus, TRAF3/4 blocks the formation of IL‐17R‐Act1‐TRAF6 complex and negatively regulating the transcription. In HeLa cells, nuclear Dbf2‐related kinase 1 (NDR1) forms a complex with TRAF3 and blocks the interaction between TRAF3 and Act1, allowing the interaction between IL‐17R and Act1 to enhance inflammation. ¹¹³

Therefore, the activation of IL‐17A signalling pathway can prolong the half‐life of mRNA of pro‐inflammatory cytokines and chemokine, through TRAF and HuR to contribute to the strong pro‐inflammatory effects of IL‐17A. The research on TRAF family and HuR is thus of significance in ischaemic stroke.

IS IL‐17B INVOLVED IN STROKE?

In 2000, the expression of IL‐17B has for the first time been detected in adult digestive tract by Northern blot but not in activated T cells. ²³ Soon after, high expression of IL‐17RB was detected in the intestines, accompanied by a large number of neutrophils infiltration in rats injected with indomethacin. ¹¹⁴ Thus, IL‐17B was proposed to be a potential pro‐inflammatory cytokine. Nowadays, IL‐17B has been proved to be prominent in inflammation and autoimmune diseases. ¹¹⁵ (Table 3) IL‐17B is highly expressed in experimental pneumonia and inducing the production of chemokine IL‐8 by activating inflammatory pathways such as PI3K/AKT, p38MAPK, ERK and NF‐κB in bronchial epithelial cells. Then, IL‐8 recruits neutrophils and promotes inflammation. ¹¹⁶ However, the expression of IL‐17RB is limited in digestive tract, lung and kidney. Whether IL‐17RB may specifically participate in the post‐stroke inflammatory response remains unknown. ¹¹⁴ At present, there is no evidence showing IL‐17B’s role in CNS, and it is still unclear whether IL‐17B is involved in the inflammatory cascade after stroke. Nonetheless, Moore et al. ¹¹⁷ detected high expression of IL‐17B in neurons, especially in grey matter projection neurons, although it is not clear whether IL‐17B is involved in the formation of neuronal axons, or the transmission of neuronal information. In a nutshell, whether IL‐17B is involved in injuring or repairing process after stroke needs elucidation.

Table 3.

IL‐17 family of cytokine and stroke

IL‐17 subtype	Length (aa)	Homology with IL‐17A (%)	Expression cells	Receptor(s)	Reception cells	Functions	Stroke	Ref.
IL‐17A	155	100	γδ T cells, NKT cells, NK cells, ILC3s, CD8+T cells and neutrophils	IL17RA/IL‐17RC	Lymphocytes, endothelial cells, astrocytes, microglia	Pro‐inflammatory, neutrophil recruitment, destroy BBB	Pro‐inflammatory, neutrophil recruitment, destroy BBB	29, 30
IL‐17B	180	29	Pancreas, small intestine, stomach, neurons, epithelial cells	IL‐17RB	Fibroblasts, THP‐1 cells, stromal cells	Pro‐inflammatory, tissue regeneration,	–	21, 112, 114
IL‐17C	197	23	Prostate, epithelial cells	IL‐17RE/IL‐17RA	Monocyte, Th17 cells	Pro‐inflammatory, neutrophil recruitment	–	21, 116, 117
IL‐17D	202	25	Muscle, brain, heart, lung, pancreas, adipose tissue	Remain unknown	Endothelial cell and myeloid progenitor	Recruitment chemokines, anti‐tumour, antiviral	–	22, 120
IL‐17E	161	17	Epithelial cells, Th2 cells, eosinophils, mast cells, brain endothelial cells	IL‐17RB/IL‐17RA	NKT cells, basophils, eosinophils, mast cells, endothelial cells, macrophages, dendritic cells, ILC2s	Induce the immune response of Th2, suppress the immune response of Th1/Th17	–	23, 121
IL‐17F	153	50	Monocytes, mast cells, basophils, mucosal epithelial cells, Th17, CD8 cells, γδ TCR+ T cells, NK, NKT, LTi	IL17RA/IL‐17RC	Non‐hematopoietic cell	Neutrophil recruitment, immunity to extracellular pathogens in inflammatory bowel disease and asthma	Pro‐inflammatory, neutrophil recruitment, destroy BBB	35

Il‐17c: “AMPLIFIER” OF Th17 SIGNAL

IL‐17C was first cloned in 2001, the same time as IL‐17B. ²³ Similar to IL‐17A and IL‐17F, IL‐17C acts as a pro‐inflammatory cytokine in the mucosal immune response. ¹¹⁸ , ¹¹⁹ After infection with Citrobacter in mice intestinal mucosa, IL‐17C was significantly increased. By binding to IL‐17RA/IL‐17RE receptor complex, IL‐17C activated NF‐κB signalling and drove the antibacterial reaction in the intestinal tract. ¹²⁰ In CNS, IL‐17C seems to play a damaging role. IL‐17RE, a specific receptor for IL‐17C, is mainly expressed in epithelial cells. ¹¹⁸ In pathogenesis of experimental autoimmune encephalitis (EAE), IL‐17RE was expressed on Th17 cells. IL‐17C/IL‐17RE complex induces the expression of IκBζ, STAT3 and RORγt through Act1, thereby inducing high expression of IL‐17A, IL‐17F and IL‐22. The process finally leads to more severe immune damage. ¹¹⁹ (Table 3) As IL‐17C amplifies the inflammatory response and promotes the progression of disease in EAE, is it applicable to all the CNS inflammatory diseases? After stroke, Th17 cells are of great importance in causing CNS inflammatory damage. Although there is no evidence that the expression of IL‐17C increases after stroke, IL‐17C could trigger the inflammatory events in CNS. This suggests that the potential pro‐inflammatory role of IL‐17C and IL‐17RE in post‐stroke.

IL‐17D: A POTENTIAL PRO‐INFLAMMATORY CYTOKINE

IL‐17D is the least studied member of the IL‐17 cytokine family, its receptor remains unknown. ¹²¹ Early studies found that IL‐17D was mainly expressed in skeletal muscle, brain and heart. IL‐17D promotes the expression of IL‐6, IL‐8 and granulocyte macrophage colony‐stimulating factor (GM‐CSF) in a dose‐dependent manner. IL‐17D may be involved in local structural damage, such as ischaemic stroke. ²⁴ IL‐17D has been found involved in cancers. Like most of the other IL‐17s, IL‐17D can induce tumour endothelial cells to produce chemokines CCL2 and MCP‐1 through the Nrf2‐IL‐17D pathway, which leads to the infiltration of NK cells and plays an anti‐tumour effect. ¹²² , ¹²³ (Table 3) As a new immune axis, Nrf2‐IL‐17D might also participate in stroke.

IL‐17E REGULATES THE BALANCE OF TH1/TH2/TH17 CELLS

IL‐25, also known as IL‐17E, was first identified from the DNA sequence of the human genome. IL‐17E is mainly expressed in epithelial cells, Th2 cells, eosinophils and mast cells ¹²⁴ and has the least homology with IL‐17A. ²⁵ During asthma, IL‐17E binds to IL‐17RB/IL‐17RA receptor complexes on Th2 cells and type 2 lymphoid cells (ILC2s) to promote the expression of IL‐4, IL‐5, IL‐13, eotaxin and type 2 immunity for airway inflammation in asthma. ¹²⁵ In contrast, IL‐17E inhibits the function of Th1 and Th17 cells. Low‐dose IL‐17E induced Th0 cells to differentiate into Th2 cells, the latter inhibited the differentiation of Th1 cells and Th17 cells by secreting IL‐4 and IL‐13 and thus plays an immune‐protective role in EAE. ¹²⁶ Moreover, exogenous administration of IL‐17E significantly inhibits CNS inflammation of EAE. ¹²⁶ (Table 3) Ischaemic stroke usually causes Th1/Th2/ Th17 cells subset imbalance and results in serious CNS inflammation and poor prognosis. ¹²⁷ As IL‐17E uniquely induces type 2 immunity to counterbalance Th17/Th1‐mediated immune pathology, it is worthy of investigating whether IL‐17E may be used to benefit stroke.

CONCLUDING REMARKS

Ischaemic stroke is an increasing threat to endanger the health and safety in modern society. Its mechanisms need to be understood. There are clear correlations between nerve damage after stroke and inflammatory response. Decoding the immune inflammatory response will improve the understanding and treatment of stroke. The IL‐17 family proteins, especially IL‐17A, are widely involved in a variety of acute and chronic inflammatory responses. After stroke, IL‐17A and IL‐17F promote the development of stroke by inducing the secretion of inflammatory factors (such as TNF‐α, IL‐6 and CXCL1), recruiting neutrophils to infiltrate into CNS and impairing the integrity of BBB. While IL‐17B/IL‐17C/IL‐17D displays pro‐inflammatory effects in plethora of diseases, their roles in stroke need further investigation as evidences are available to show IL‐17B/IL‐17C/IL‐17D may be damaging in stroke. On the contrary, IL‐17E plays an immune‐protective role by balancing Th2 vs Th1/Th17 cell functions. In conclusion, the IL‐17 family proteins are prominent and pleiotropic to modulate both pro‐ and anti‐inflammatory responses and may be exceptional therapeutic targets for treating stroke.

AUTHOR CONTRIBUTIONS

Qiaohui Zhang and Yibo Tang conceived and wrote the manuscript; Yan Liao, Zhenquan Liu polished the manuscript; Yajie Dai, Yunxin Li and Yue Li provided many comments for the manuscript.

DISCLOSURES

The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT

The data used to support this review are included within the article.